[haiku-development] Re: APM and shutdown

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

could not have put it better myself.

yet i think i was told that 'to implement these changes would
significantly DELAY the R1 release date.'

i found it ran fine otherwise under leopard intel and qemu (Q). i have
yet to test on ppc, should yeild same result on my tower under ubuntu.

An INTEL is an INTEL no matter what OS you use.

I would RECOMMEND this and a bootable image or drive as 'priorities'.
Why QEMU a dev enviroment when you can boot it native? QEMU uses up more
resources to emulate than it would otherwise take to run the OS
anyway.Im at 60% load under either QEMU or VPC on my 600mhz ibook.Any
more load to it and the master OS grinds to a halt. Gives you an idea
what I'm running at. (Approx. 350+ Mhz.Rather slow.)

Most VESA resolutions are accepted by a majority of cards nowadays
anyway, and this is what by default Ubuntu goes to if it for some reason
cant (wont in my case on the laptop) find your video card chipset.
Sound, well, this is a dev enviroment, I'll leave that up to you.

i would say that most everything else is already there, just needs some
tweaking. Maybe NET needs some help, but if you're a dev, (who cares)
hell as a user, I've delt with less than perfect OS setups. NET is good,
but for dev enviroment, not necessarily necessary. By default, NET with
OSX86 Intel Leopard (not apple's default, the spinoff version) is
usually NOT working, you have to tweak and poke at it to get it working.
For production, though, you NEED it. Maybe R2?

Good job so far.
- -----

Andreas Färber wrote:
> Hello,
> 
> Through r26937 I noticed that the kernel settings file has an option to
> enable APM. Enabling it works fine in QEMU (gcc4), in the form that QEMU
> exits when shutdown is finished. Same for real hardware (gcc2). A ticket
> for the shutdown feature was recently turned down as "invalid":
> http://dev.haiku-os.org/ticket/2179
> Is there still a reason for not turning on APM by default? Axel reported
> it not working for one out of four of his machines three years ago
> (r16152) and Jack wrote in #2179 it panicked for him in the past, so
> maybe that could be re-tested to see if it's fixed now or specific to
> certain hardware?
> 
> Anyhow, having this enabled and working for the Alpha would seem like a
> good idea to me, since not shutting down properly can be perceived as
> inconvenient for people used to it from Windows, Linux etc. It would
> give a bad final impression, unless the user restarts. Not to mention
> that it saves energy.
> 
> We have had reports of certain pieces of hardware leading to errors
> before, but to my knowledge that has not lead to, e.g., IDE/SATA being
> disabled on all machines just because of a JMicron controller not
> working somewhere.
> If we don't test APM because we're unaware of its existence or because
> tickets indicating such an unresolved problem get closed (and are thus
> no longer listed), then we are unlikely to find out where it fails and
> where it doesn't, and hence will not see much progress there...
> 
> Andreas
> 
> 
> 
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFIoecSZWyv5fvyGz0RAl8pAJ9VxFhvm9mv2PrCVl6O8P5m7Fm/3wCgoUOK
HcfbRLLAYoXoXV9pvUSNzkk=
=iGz9
-----END PGP SIGNATURE-----


Other related posts: